Prostate cancer is one of the most common malignant diseases for which health-care intervention is sought worldwide. In many developed countries it is the most common non-cutaneous malignant disease. In US only, estimated new cases 2008 are about 186,320, and the death rate estimated to be about 28,660 for 2008 (US National Cancer Institute, estimated statistics for Prostate cancer, www.cancer.gov/cancertopics/types/prostate).
Men with prostate cancer have mostly no symptoms at all, especially in the early stages. Some symptoms that may indicate prostate cancer include difficult and painful urination, frequent urination, incomplete emptying of the bladder, decrease force of urine stream, blood in the urine, hip and back pain. When the cancer has spread to other parts of the body, symptoms may include bone pain, weight loss, anemia, kidney failure, weakness or paralysis caused by the spinal cord.
There are many types of prostate cancer and it is often present in many different parts of the prostate. The precursor to prostate cancer is known as prostatic intraepithelial neoplasia (PIN), this is also found in many different locations within the prostate.
Although there are many different kinds of prostate cancer the vast majority (around 95%) are of the type known as adenocarcinoma. As this is the most wide spread from it has become synonymous with the term prostrate cancer.
The current preferred method for diagnosing early prostate cancer is prostatic needle biopsy since it has a low morbidity and provides specific information on the grade and extent of the tumour.
The most common site of origin of prostate cancer is in the peripheral zone (the main glandular zone of the prostate). The term adenocarcinoma originates from Adeno meaning ‘pertaining to a gland’, whilst Carcinoma relates to a cancer that develops in epithelial cells.
Available markers for diagnosing prostate cancer are inferior and the markers available have serious limitations relating to specificity. This results in less objective results by attribution or assignment bias which limits the accuracy of the resulting information.
Prostate-specific antigen, PSA, is still the main diagnostic tool despite its serious limitations, while studies of new markers are being performed and reported continuously.
PSA is a protein produced by both normal and cancerous prostate cells. A high level of PSA can be a sign of cancer, but the PSA level can also be raised in prostate conditions that are not cancer, i.e. they are benign, or if you have an infection. PSA is analyzed in blood samples. Since it affects the conclusion(s) drawn from the measured levels, one wants to rule out infections like e.g. a urine infection before carrying out a test.
Further, there is no PSA reading that is considered ‘normal’. The reading varies from man to man and the normal level increases as you get older. As a rule of thumb, the higher the level of PSA, the more likely it is to be cancer. Sometimes a cancer may be diagnosed in a man with a ‘normal’ PSA reading. But usually, the higher the reading, the more likely it is to be cancer. There is thus a high uncertainty and certainly an attribution of subjectivity in diagnosing prostate cancer using PSA.
New studies emerging for other markers useful for diagnosing prostatic cancer are still subjected to attribution or assignment bias (depending on the pathologists experience and awareness) with more or less arbitrary chosen cut-off values used as standard values to decide the outcome of the decisive test where an assumption is made that no cancers are present below that cut of. The assumption may thus leading to wrong findings and, subsequently, wrong diagnosis for the patient.
US 2005/0186642 provides regents with primary antibodies and a detection system using antibody cocktails. Antibody cocktails are exemplified in the application with compositions comprising antibodies binding to i) high-molecular weight cytokeratin (HMWCK) (mouse antibody)+medium weight cytokeratin MWCK (rat antibody), and a composition comprising antibodies binding to ii) α-metyhylacyl-CoA-racemase (AMCAR) (rat antibody)+HMWCK+p63 (both mouse antibodies) (PIN cocktail).
Schuyler et al (Am. J. Clin. Pathol. 2004, 121:220-225) provides an antibody cocktail with antibodies binding to p63 and AMCAR in evaluation of prostate biopsy specimens (a PIN cocktail).
Molinié et al. (Modern Pathology, 2004, 17, 1180-1190) provides staining and evaluation of prostate specimen from surgical pathological files using antibodies binding to high molecular weight CK 5/6 or a cocktail with antibodies binding to p63 and AMCAR (AMCAR/p63, PIN cocktail).
Srigley, J., (Modern pathology, 2004, 17:328-348) provides immunohistochemical markers for benign mimickers of prostatic adenocarcinoma in diagnosis, and problems related thereto. Antibodies binding to HMWCK, cytokeratin 5 and 6 (CK5/6) and p63 are mentioned as basal cell markers and AMCAR/p63 is used as a PIN cocktail.
p63 is a nuclear protein selectively expressed in the basal cell compartment of a variety of epithelial cells. Normal prostate glands show selective p63 nuclear expression in basal cells and p63 is consistently undetectable in prostate cancer (Signoretti et al., 2000, Am. J. Pathol., Vol. 157, pg. 1768-1775). Antibodies raised to p63 thus give nuclear staining.
Nuclear staining is attributed to some inherent technical problems giving rise to an inconsistency and even lack of repeatability due to technical reasons. The fixation and permeabilisation steps always affect the availability of a proteins (i.e. the specific antigen) and thus, nuclear staining are more dependent thereon than cell-surface or cytoplasmic proteins due to the permeabilisation of the nuclear envelope. Further, even and exact cuttings as such as well as thickness thereof and spatial localisation of the cut in relation to the orientation of the cell also affect the availability of nuclear proteins, such as p63, to a higher degree than for cell-surface or cytoplasmic proteins. Also, a dividing cell may give rise to a diffuse staining—if any—compared to a non-dividing cell.
Further to be mentioned is that p63 is expressed in most, but not all, basal cells surrounding the prostate glands which further gives a degree of uncertainty when analysing staining of normal and possible prostate cancer tissue samples (Signoretti et al., 2000, Am. J. Pathol., Vol. 157, pg. 1768-1775).
Further, it is known in the art that stainings of the p63 antigen is sensitive to storage over time and a time course study has shown a progressive decline in the p63 intensity score with time (Effect of Storage on p63 Immunohistochemistry: A Time-course Study. Appl Immunohistochem Mol Morphol. 2009 January; 17(1):68-71).
Inconsistency in staining of the nuclear protein p63 will, of course, affect the conclusions drawn from the detection of this protein in the samples analysed. Thus, affecting the conclusions, even leading to a wrong conclusion, will affect the further diagnosis or prognosis of the patient and even the future care, treatment and even outcome of the individual. A wrong diagnosis or prognosis is thus highly un-appreciated both by patient and the hospital care system.
It is thus highly desirable to have a more reliable, but still, an equally easy detection of prostate cancer that further avoids the use of staining nuclear proteins such as e.g. p63 frequently used in new studies emerging on antibody-cocktails for detection and diagnosing of prostate cancer.
There is thus an urgent need to find better diagnostic and prognostic markers, means and methods when diagnosing and prognosing prostatic cancer in a simple and reliable way, as well as less biased means to perform an accurate and less biased method or assay for detecting prostatic cancer. Accordingly, the present invention seeks to provide means and methods to perform accurate and less biased diagnostic assays, in a simple and efficient way for routine testing when diagnosing or prognosing prostatic cancer.